Meta Puts AI Assistant Muse at the Centre of Its Future as New Shares Support Expansion
Meta Platforms is reorganising its artificial intelligence strategy around Muse, its new personal AI agent, as the Facebook and Instagram owner prepares to issue a new class of non-voting shares to support its costly expansion. The strategy signals a shift towards AI-powered services that can perform tasks for users, rather than simply answer questions, while giving Meta additional financial flexibility to compete in an increasingly crowded market.
The announcement came alongside the company's third-quarter results, which showed stronger-than-expected revenue and rapid early adoption of Muse. However, the growing cost of developing AI models, building data centres and recruiting specialised talent remains a significant challenge for the technology giant.
Muse Emerges as a Core Part of Meta's AI Strategy
Meta introduced Muse in September as a personal AI agent designed to help users complete tasks across connected applications. Unlike conventional chatbots that primarily respond to prompts, Muse is intended to carry out activities such as organising emails, planning trips, managing schedules and assisting with online purchases.
The service is powered by Meta's Muse Spark model and operates through a dedicated virtual environment. Users can decide which services it connects to and what permissions it receives, while certain sensitive actions require approval.
Chief Executive Mark Zuckerberg sees the product as a potential gateway to a much broader AI ecosystem. According to the company's latest update, more than 10 million people in the United States and Canada had established Muse accounts within roughly six weeks of its launch.
Meta hopes that wider adoption will create a network effect: as more individuals and businesses use the assistant, it could become more useful and open opportunities for additional services and commercial transactions.
New Class of Shares to Help Fund Expansion
Meta plans to issue a new class of non-voting Class C common shares beginning in January 2027. The shares are intended to provide additional flexibility for acquisitions, employee compensation and other strategic investments without weakening Zuckerberg's control over the company.
The structure is effectively designed to produce a two-for-one stock split, although investors should distinguish the new share class and its implications from a conventional cash-raising share sale. The company's stated objective is to expand its financing options while preserving its existing control arrangements.
The move comes as Meta commits enormous resources to AI infrastructure, including advanced computing chips, data centres and the development of increasingly capable models. Its capital expenditure forecast for 2026 now stands at between $135 billion and $145 billion, with the lower end raised from its previous outlook.
Such spending reflects the scale of the competition Meta faces from leading AI developers and other technology companies seeking to build consumer and business AI products.
Strong Revenue, but AI Costs Are Rising
Meta reported third-quarter revenue of $64.60 billion, exceeding the $63.37 billion expected by analysts surveyed by LSEG. The company also forecast fourth-quarter revenue of between $73 billion and $76 billion, indicating that its core advertising business continues to provide substantial financial support for its expansion.
However, its cash position highlights the pressure created by investment. Meta recorded a free cash flow deficit of $2.12 billion in the third quarter, compared with positive free cash flow of $10.63 billion a year earlier.
The company's financial outlook therefore depends on balancing two priorities: maintaining growth in advertising across Facebook and Instagram while developing AI products that can eventually generate meaningful additional revenue.
Chief Financial Officer Susan Li expects AI businesses to begin adding incremental revenue in 2027 and become a more meaningful source of diversification over time. Meta has also indicated that it intends to monetise Muse sooner than it typically would with a new product because the service is expensive to operate.
Reality Labs Shifts More Attention Towards AI
Meta is also adjusting its investment priorities within Reality Labs, the division responsible for virtual reality, augmented reality and related technologies. The company has spent heavily on its metaverse ambitions, but it is now directing more resources towards AI-related opportunities.
Meta expects Reality Labs' losses to narrow to around $15 billion next year as investment is reallocated towards priority areas, primarily artificial intelligence. The change does not mean that Meta is abandoning its work on immersive technologies; rather, it indicates a stronger focus on areas management believes can deliver more immediate strategic or commercial value.
AI assistants could also become increasingly relevant to smart glasses and other connected devices, giving Meta additional ways to integrate its technology into users' daily routines.
Privacy and Reliability Remain Major Challenges
Despite Muse's early adoption, its expansion raises important questions about privacy, security and reliability. An assistant capable of interacting with email, calendars, shopping services and payment tools may handle sensitive information and perform actions that have real-world consequences.
Meta says Muse includes safeguards such as permission controls, a separate security-monitoring agent and approval requirements for certain sensitive actions. Users can also limit access to connected applications.
Nevertheless, the broader AI industry continues to face concerns about inaccurate responses, inappropriate access to information and the possibility that automated systems could take unintended actions. Greater adoption could increase scrutiny of how personal data is handled and how users can review or reverse an agent's decisions.
Meta also faces wider legal and regulatory pressures. These issues could increase compliance costs and influence public trust as the company expands AI services that interact more directly with personal information.
What the Strategy Means for Meta's Future
Meta's decision to put Muse at the centre of its AI plans represents an attempt to turn substantial technology spending into a new long-term business opportunity. A successful assistant could strengthen user engagement, support subscriptions and transactions, and create additional commercial opportunities beyond advertising.
The risks are equally significant. Infrastructure spending could continue to weigh on cash flow, rivals may develop competing products, and privacy concerns could slow adoption. The new share structure may offer financial flexibility, but it does not remove the need for Meta to demonstrate that its investments can generate sustainable returns.
For investors, the central question is whether Muse and Meta's wider AI portfolio can eventually deliver enough additional revenue to justify the enormous cost of development. The company's latest results provide an encouraging starting point, but the longer-term outcome will depend on user retention, reliable performance, responsible data handling and successful monetisation.
Meta is betting that personal AI agents will become a major part of how people use technology. Its next challenge is to prove that this vision can become a durable and profitable business.